The rabbit is the most appropriate animal model in which to test candidate immunogens for beneficial and detrimental effects on the pathogenesis of tuberculosis. Mice show relatively low tuberculin sensitivity, show relatively little caseous necrosis, and never produce liquefaction and cavities, but mice are a good model for cell-mediated immunity (CMI). Guinea pigs develop rather poor CMI, only occasionally develop cavities, and are much more susceptible to tuberculosis than human beings. In rabbits, lesions produced by the intradermal injection of the candidate vaccine itself will first be evaluated. Then, we shall evaluate the beneficial (and detrimental) effects of candidate vaccines (a) on dermal BCG lesions, (b) on pulmonary lesions produced by aerosols of virulent human-type tubercle bacilli, and (c) on cavity for formation produced by aerosols of virulent bovine-type tubercle bacilli (which are equally virulent in human beings). Periodic tuberculin tests will also be made. Each type of lesion is evaluated for: (a) rate of healing or progression; (b) the amount of caseous necrosis, liquefaction and cavity formation; (c) the number of activated macrophages present by means of our B- galactosidase and acid phosphatase histochemical tests; and (d) the number of bacilli present by acid-fast staining and by culturing on solid egg media. When indicated, immunohistochemical tests for types of T cells and various cytokines will be performed, as well as "in situ hybridizations" for cytokine mRNA. A vaccine's ability to reduce the number of primary tbc lesions in rabbits exposed by aerosol to virulent (human-type) bacilli is the most appropriate test existing to measure the vaccine's ability to prevent clinical tuberculosis. The most ideal vaccine should not produce tuberculin sensitivity, and, following challenge with virulent bacilli, should reduce the amount of caseation, liquefaction and cavity formation.